[tuples] fix location problems with optimized GIMPLE_CONDs

Diego Novillo dnovillo@google.com
Tue Jul 22 22:59:00 GMT 2008


On Tue, Jul 22, 2008 at 15:52, Aldy Hernandez <aldyh@redhat.com> wrote:

> I'll commit once the tests are finished, unless someone has some show
> stopping advice.

I think this is fine.

There doesn't seem to be a clear place where gimple_set_location() is
initialized for every new statement.  Are you relying on PLFs starting
up set to zero after a tuple is created?


Diego.



More information about the Gcc-patches mailing list